2010 NBA Playoffs: Projecting The Lakers-Suns Series

May 13, 2010 - by Tom Federico

With both the Lakers and the Suns coming off series sweeps, the Western Conference finals feature a battle between two red hot teams. The Lakers have home court advantage, but our predictive power ratings see the Suns as the slightly higher performing team overall, both over the course of the regular season plus playoffs to date, and in recent games.

In recent posts, we’ve made numerous mentions of the potential downsides of using a power ratings based methodology for predicting the NBA playoffs. This methodology made the wrong call for the Lakers-Jazz conference semifinals series, but is 10-1 picking playoff winners so far in 2010. All of our 2010 NBA Playoffs predictions are posted.

In short, if you believe the Lakers’ past performance is not a great predictor of their likely performance in this series, and that Kobe and crew can “step it up” much better, relatively, than the Suns can, then take the following analysis with a grain of salt. Despite the inherent limitations of the data available to us, we still favor going with the numbers over making guesses about who can step it up better or who wants it the most.

We see this series as a toss up, with the Suns, despite not having home court advantage, having about a 52% of winning overall. That’s certainly within forecasting margins of error, so let’s just call it 50/50.

This prediction contrasts quite starkly with the current Vegas odds for the series: Lakers -330, Suns +289. Those odds imply that the Lakers will win this series about 75% of the time, not 50% of the time as we project.

That difference could be the result of several factors. Here are two possibilities:

1) It’s well known that bettors typically pay a premium to bet on popular teams, and the Lakers are a prime example. LA being so close to Vegas also supposedly factors into this situation, and coming off an unexpected sweep, public perception of the Lakers just got a huge boost.

2) A power ratings based prediction methodology indeed overvalues the relevance of LA’s regular season results during playoff time, much more so than it does for the Jazz.

Below are the preliminary results from our series simulations. We still have a little tweaking to do, but nothing too major should change. Based on the initial results of simulating the Lakers-Suns series 10,000 times:

– The series goes 4 games 1,160 times; Phoenix wins about 60% of them
– The series goes 5 games 2,450 times; LA wins about 52% of them
– The series goes 6 games 3,226 times; Phoenix wins about 62% of them
– The series goes 7 games 3,164 times; LA wins about 56% of them

Based on this analysis, our pick right now would be Phoenix in 6 games, which the numbers indicate has about a 20% chance of happening. Other observations:

– The second most likely result is the Lakers in 7 games (17-18% chance)
– Phoenix has about a 7% chance to sweep the Lakers
– The Lakers have about a 5% chance to sweep Phoenix

The Orlando vs. Cleveland/Boston series should be very interesting to project. That analysis will be coming soon.